| *** promethe- is now known as prometheanfire | 01:30 | |
| opendevreview | Merged openstack/openstack-zuul-jobs master: Drop lib-forward-testing template https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/911088 | 06:15 |
|---|---|---|
| opendevreview | daniel.pawlik proposed openstack/ci-log-processing master: Update base container image to Centos 9 Stream https://review.opendev.org/c/openstack/ci-log-processing/+/911106 | 08:48 |
| opendevreview | daniel.pawlik proposed openstack/ci-log-processing master: Set the base_python version to 3.11 https://review.opendev.org/c/openstack/ci-log-processing/+/906637 | 08:48 |
| dpawlik | fungi, clarkb1: hey, do you know why "Web Listing Disabled" was done for logs dir? - for example https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_e9c/910700/3/check/cinder-plugin-ceph-tempest/ | 09:51 |
| dpawlik | Is it because of the logscraper ? | 09:51 |
| frickler | dwhite4: that message usually appears when the logs have been removed because they were too old, let me check | 09:53 |
| frickler | ehm, sorry dwhite4 meant dpawlik | 09:53 |
| tonyb | `The owner of this web site has set X-Container-Meta-Web-Index: index.html. However, this file is not found.` That seems strange | 09:54 |
| frickler | ah, no, the logs are under https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_e9c/910700/3/check/cinder-plugin-ceph-tempest/e9c62c5/ | 09:55 |
| frickler | and since this is object storage, there is no real filesystem tree behind it, so truncating the path doesn't work | 09:56 |
| tonyb | Ah okay | 09:56 |
| tonyb | I see, zuul links to the full (storage) URL | 09:57 |
| frickler | so for the real log directory, an index.html file was generated and uploaded by zuul. for any truncations for the path, there is not matching index file and the above error happens | 09:58 |
| frickler | modulo grammar | 09:58 |
| tonyb | Makes sense | 09:58 |
| frickler | dpawlik: does that help you? | 09:59 |
| opendevreview | daniel.pawlik proposed openstack/ci-log-processing master: Add capability with new Zuul https://review.opendev.org/c/openstack/ci-log-processing/+/911423 | 10:55 |
| opendevreview | daniel.pawlik proposed openstack/ci-log-processing master: Remove unimportant fields from buildinfo https://review.opendev.org/c/openstack/ci-log-processing/+/911424 | 10:57 |
| opendevreview | daniel.pawlik proposed openstack/ci-log-processing master: Update base container image to Centos 9 Stream https://review.opendev.org/c/openstack/ci-log-processing/+/911106 | 10:58 |
| opendevreview | daniel.pawlik proposed openstack/ci-log-processing master: Set the base_python version to 3.11 https://review.opendev.org/c/openstack/ci-log-processing/+/906637 | 10:58 |
| opendevreview | daniel.pawlik proposed openstack/ci-log-processing master: Add capability with new Zuul https://review.opendev.org/c/openstack/ci-log-processing/+/911423 | 11:26 |
| fungi | dpawlik: put another way, that swift service doesn't have magic autoindexing like you'd expect from apache mod_autoindex on a more typical webserver | 12:56 |
| opendevreview | Jonathan Rosser proposed openstack/project-config master: Implement ironic-unmaintained-core group https://review.opendev.org/c/openstack/project-config/+/911576 | 13:12 |
| opendevreview | Jonathan Rosser proposed openstack/project-config master: Implement openstack-ansible-unmaintained-core group https://review.opendev.org/c/openstack/project-config/+/911576 | 13:13 |
| frickler | jrosser_: ^^ gerrit acls is one of the very few locations where tabs instead of spaces are actually required | 13:22 |
| jrosser_ | doh | 13:22 |
| opendevreview | Jonathan Rosser proposed openstack/project-config master: Implement openstack-ansible-unmaintained-core group https://review.opendev.org/c/openstack/project-config/+/911576 | 13:23 |
| *** tosky_ is now known as tosky | 13:57 | |
| opendevreview | daniel.pawlik proposed openstack/ci-log-processing master: Add capability with new Zuul; disable log-gearman CI job https://review.opendev.org/c/openstack/ci-log-processing/+/911423 | 14:43 |
| *** dansmith_ is now known as dansmith | 15:02 | |
| opendevreview | Jeremy Stanley proposed openstack/project-config master: Try switching Rackspace DFW to an API key https://review.opendev.org/c/openstack/project-config/+/911381 | 15:06 |
| opendevreview | daniel.pawlik proposed openstack/ci-log-processing master: Remove log gearman functionality https://review.opendev.org/c/openstack/ci-log-processing/+/911590 | 15:16 |
| tbachman | Hi fungi! I'm back with a follow-up question | 15:19 |
| tbachman | I should have caught the upper-constraints issue before - my apologies on that (and thanks for pointing that out!) | 15:19 |
| tbachman | But now I have a weird issue. | 15:19 |
| tbachman | This patch is an attempt to align with the upstream unmaintained branches for yoga: | 15:20 |
| tbachman | https://review.opendev.org/c/x/group-based-policy/+/911385 | 15:20 |
| tbachman | The part that's weird is that the py36 and py38 jobs pass, but the py39 doesn't. The reason is that the py39 job is picking up the wrong version of upstream neutron (23.1 instead of 20.5) | 15:22 |
| tbachman | I'm not sure why the py36/38 and py39 jobs would be different here | 15:22 |
| fungi | tbachman: for one thing, it looks like you don't have openstack/requirements checkout overridden with an openstack-tox-py39 variant in your .zuul.yaml, but you do for 36 and 38 | 15:23 |
| tbachman | Yeah - but I did try adding that, but ended up with a different failure/issue. | 15:24 |
| tbachman | It was timing out | 15:24 |
| * tbachman goes to get link | 15:24 | |
| fungi | tbachman: probably another problem you ran into with py39 is if you tried to tell it to run on ubuntu-bionic | 15:24 |
| fungi | since ubuntu-bionic doesn't have a python3.9 package | 15:25 |
| tbachman | ah | 15:25 |
| tbachman | :( | 15:25 |
| tbachman | sorry | 15:25 |
| tbachman | okay - will pick a different node then. Thanks! | 15:25 |
| fungi | yw. worth looking at https://governance.openstack.org/tc/reference/runtimes/yoga.html and also the upstream job definitions when working out what platforms are expected to work | 15:26 |
| tbachman | got it - thanks again! | 15:26 |
| fungi | tbachman: https://opendev.org/openstack/openstack-zuul-jobs/src/branch/master/zuul.d/jobs.yaml#L242 indicates upstream definition for openstack-tox-py39 uses the ubuntu-focal nodeset when run for unmaintained/yoga | 15:28 |
| fungi | so if you don't explicitly override the nodeset in your variant, then it should just inherit ubuntu-focal for that | 15:29 |
| fungi | technically it's matching (among other patterns) ^.*/yoga so should end up selected for both stable/yoga and unmaintained/yoga | 15:30 |
| opendevreview | daniel.pawlik proposed openstack/ci-log-processing master: Remove log gearman functionality https://review.opendev.org/c/openstack/ci-log-processing/+/911590 | 15:31 |
| fungi | gmann: elodilles: that discussion got me looking, i think we missed broadening the negation for some job branch matchers, since this is excluding stable/yoga but not unmaintained/yoga so zuul is going to potentially choose both it and the next definition: https://opendev.org/openstack/openstack-zuul-jobs/src/branch/master/zuul.d/jobs.yaml#L232-L233 | 15:32 |
| fungi | as a result, it's unclear whether it will still end up forcing tox<4 | 15:33 |
| fungi | ideally the two matchers should be inverses of one another | 15:34 |
| fungi | seeing the same problem in a bunch of other variants in that same config | 15:35 |
| frickler | yes, needs updating from victoria through to yoga by now | 16:01 |
| tbachman | fungi: yeah - we used focal in the latest iteration, which passed the gate. Thanks again for your help! | 16:24 |
| opendevreview | daniel.pawlik proposed openstack/ci-log-processing master: Remove log gearman functionality https://review.opendev.org/c/openstack/ci-log-processing/+/911590 | 16:28 |
| fungi | tbachman: great to hear! | 16:28 |
| elodilles | fungi: i'm quite sure that many job defs will need some refactor due to stable/ -> unmaintained/ change :/ some jobs (non-devstack based ones) are currently running on unmaintained/ branches. so yes, another TODO around the unmaintained changes | 16:36 |
| *** clarkb1 is now known as clarkb | 16:55 | |
| gmann | fungi: elodilles yes, we need to correct a lot of variant for v-y unmaintained renaming | 18:09 |
| *** kopecmartin_ is now known as kopecmartin | 22:50 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!